Nuclear PSAs

It was stated earlier that PSAs for nuclear and chemical processes differ by preceding a chemical PSA with a scoping analysis that is omitted for a nuclear PSA. This is not true. WASH-740 performed this purpose for a hypothetical plant. 

Nuclear PSAs tend to accept a set of initiators such as the WASH-1400 set or the EPRI set and proceed with little introspection as to completeness of the set for a specific plant. A well-known omission is sabotage. In general, earthquake and fire is treated poorly tornado is usually not addressed nor are other aspects of adverse weather such as freezing and ice storms which may have common cause potential. The record has not been too good. After the Browns Ferry fire, it was claimed that such was addressed in WASH-1400. If so, attention was only brought after the fact. The TMI-2 sequence was not addressed in the PSAs to an extent that action was taken to prevent it. 

Nuclear PSAs contain considerable uncertainty associated with the physical and chemical processes involved in core degradation, movement of the molten core in the reactor vessel, on the containment floor, and the response of the containment to the stresses placed upon it. The current models of these processes need refinement and validation. Because the geometry is greatly changed by small perturbations after degradation has commenced, it is not clear that the phenomcn.i can be treated. 

Unfortunately chemical process plant PSA have no requirements for public disclosure, consequently examples of their PSAs are hard to find. However the following are some examples from the open literature. They are similar to nuclear PSAs except less elaborate with more empahsis on consequences than on probabilities. 

A technique called probabiUstic safety assessment (PSA) has been developed to analy2e complex systems and to aid in assuring safe nuclear power plant operation. PSA, which had its origin in a project sponsored by the U.S. Atomic Energy Commission, is a formali2ed identification of potential events and consequences lea ding to an estimate of risk of accident. Discovery of weaknesses in the plant allows for corrective action. 

One of the products of a nuclear power plant PSA is a list of plant responses to initiating events (accident starters) and the sequences of events that could follow. By evaluating the significance of the identified risk contributors, it is possible to identify the high-risk accident. sequences and take actions to mitigate them. 

Perhaps the most readily perceived type of presentation is a "pie" chart of the major contributors as determined in the Zion and Arkansas Nuclear One (ANO)-l) PSAs and presented in Figure 1.4.3-6. 

Department of Defense - requires that a PSA be performed according to MIL-STD-882A for any major acllvity or undertaking, e.g., analyses of the transportation of nuclear weapons and deactivatioti of chemical weapons. 

There are no PSA requirements for licensing the current generation of nuclear power plants, i (PRA to the NRC) has been a tool for regulatory decisions b]... 

CFR52 requires a PSA for the next generation of nuclear power plants,... 

Figures 2.8-1 and 2.8-2 from Vesely et al. (1983) show how the RAWR and RRWR measures vary for two PWR nuclear power plants (Oconee, and Sequoyah) based on their PSAs. This variation may be explained on the basis of different manufacturers of the plants (Babcock Wilcox and Westinghouse, respectively)...

FMEA is particularly suited for root cause analysis and is quite useful for environmental qualification and aging analysis. It is extensively used in the aerospace and nuclear ]iowei indiistrii-s but seldom used in PSAs, Possibly one reason for this is that FMEA, like parts count. ,s not chrectlv suita lundant systems such as those that occur in nuclear power plants Table i 4... 

Table 3.6-5 presents code usage in the PSAs ot some nuclear power plants, and Table. 3 6 6 sur eys codes used in major... 

RISKMAN is an integrated Microsoft Windows , personal computer software system for [H. i forming quantitative risk analysis. Used for PSAs for aerospace, nuclear power, and chemical [iroccsses, it has five main modules Data Analysis, Systems Analysis, External Events Analysis, Event Tree Analysis, and Important Sequences. There are also modules for software system maintenance, backup, restoration, software updates, printer font, and page control. PEG has also integrated the fault tree programs CAFTA, SETS, NRCCUT, and IRRAS into RISKMAN. 

Table 4.3-3 from Joksimovich et al. (1983) presents the data sources used in preparing the Big Rock Point and Zion nuclear power plant PSAs. It is seen that both PSAs used plant records extensively. 

This chapter overviews the techniques for incorporating external events into a PSA. The discussion was primarily aimed at nuclear power plants but is equally applicable to chemical process plants. The types of external events discussed were earthquakes, fires and floods. Notably absent were severe winds and tornados. Tornados are analyzed as missiles impacting the structures and causing common-cause failures of systems (EPRINP-768). Missile propagation and the resulting damage is a specialized subject usually solved with computer codes. 

The preceding overviewed the operation and engineered safety features of current and advanced LWRs. Before preceding to describe how PSA is performed on nuclear power plants, two accidents are described that have profoundly affected the industiy... 

While event trees are not so common in chemical plant PSA they are extensively used to analyze nuclear accidents, but before beginning the analysis, preliminaries that are needed are plant tamiliarity and initiator selection... 

Worker risk is not usually included in nuclear power plant PSAs. If it were, it would be conducted similarly to chemical plant PSAs. 

CR-2300 (PRA Procedures Guide) with actual expenditures (Joksim-ovich, 1983). In this comparison of analyses. Grand Gulf is a level 1 PSA Arkansas Nuclear 1 (ANO-1) is an IREP level 2 PSA Big Rock Point (BRP) is a level 3 PSA with limited treatment of external events but thorough in consideration of environmental effects on equipment. The Zion PSA thoroughly treats both... 

It is more common for a complete PSA to take 16 to 24 months with several rnuiilhs tor preparation, review, and revision of the final report. The final report for a level 3 nuclear plant PSA, includes an analysis of external events, in several large volumes. Completeness and consistency in such a large document requires several months of team leadership and selected analysts. Given these resources, it may be possible to complete the technical analyses for a Level 1 PSA in a year or less, but the final report will take several more months to prepare. 

For historical and regulatory reasons, a PSA of a nuclear power plant begins with the system analysis to determine the ways that an upset condition could occur, its probability... 

Nevertheless, WASH-740 indicated the worst-case hazard of nuclear power, and the need for detailed PSAs of nuclear power plants. Since plants proposed for construction are sufficiently like the hypothetical plant in terms of fission products subsequent scoping analyses are not needed before the detailed PSA. 

Table 7.3,2-3 lists investigations to be conducted and documented for a detailed chemical plant PSA. The steps are similar to those required for a nuclear plant PSA except the hazards are more varied, and dispersed regarding concentration and location. Many of the steps previously described in Section 6.3.2 are applicable for the chemical PSA a-, well. 

Regulatory Guide 1.113 provides procedures for estimating the aquatic transport of accidental nd routine releases from nuclear reactors. Because these methods are complex and have not been included in PSAs, they are not discussed. 

The nuclear equipment failure rate database has not changed markedly since the RSS and chemical process data contains information for non-chemical process equipment in a more benign environment. Uncertainty in the database results from the statistical sample, heterogeneity, incompleteness, and unrepresentative environment, operation, and maintenance. Some PSA.s use extensive studies of plant-specific data to augment the generic database by Bayesian methods and others do not. No standard guidance is available for when to use which and the improvement in accuracy that is achieved thereby. Improvements in the database and in the treatment of data requires, uhstaiui.il indu.sinal support but it is expensive. 

The primary motivation of PSAs is to assess the risk of the plant to the public. The immediate purpose of the RSS was to support the Price-Anderson hearings on liability insurance (i.e., assess the financial exposure of a nuclear power reactor operator) a purpose which, even today, is beyond PSA technology. However, PSA is sufficiently precise to provide relative risk comparisons of reactor designs and sites. These uses of PSA were presented at the Indian Point hearings, and in defense of Shoreham. The PSAs for the high-population-zone plants (Limerick, Zion, and Indian Point) were prepared to show that specific features of these plants compensate for the higher population density relative to plants studied in the RSS. 

The RSS was used by pro- and anti-nuclear power advocates in the Proposition 15 (nuclear power ban) voting in California the Barsebek and Ringhals PSAs were prepared as part of the nuclear controversy in Sweden the German Risk Study, 1981, had similar applications as did the Sizewell-B studies in Great Britain,... 

---